基于可变权重的威布尔–广义帕累托模型的巨灾损失分布拟合

Catastrophe Loss Distribution Fitting Based on Variable Weighted Weibull-Generalized Pareto Model

摘要: 我国是世界上台风灾害频发的国家之一，由此造成的人员伤员和财产损失逐年上升，影响经济发展。本文以台风风暴潮数据为例，采用1989年~2021年我国沿海地区台风风暴潮直接经济损失数据作为样本，对数据进行预处理后同时引入以威布尔分布、帕累托分布和广义帕累托分布为基础构建的组合分布模型进行拟合，结果表明：可变权重的威布尔–广义帕累托模型的拟合效果最好，并以此为依据为我国台风巨灾损失提供分析方法。

Abstract: China is one of the countries in the world with frequent typhoon disasters, and the resulting casualties and property losses have increased year by year, affecting economic development. Taking typhoon storm surge data as an example, this paper uses the direct economic loss data of typhoon storm surge in coastal areas of China from 1989 to 2021 as a sample, preprocesses the data, and introduces the combined distribution model constructed on the basis of Weibull distribution, Pareto distribution and generalized Pareto distribution for fitting, and the results show that the Weibull-Generalized Pareto model with variable weight has the best fitting effect, and provides an analysis method for typhoon catastrophe loss in China based on this.